© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 67932
    Prev
    Next
    INDEX
    Explanations

    evaluating candidates and criteria

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron flags language about evaluating or testing knowledge—especially references to assessments (e.g. “tests,” “purpose,” “estimate,” “discriminate”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Molo
    -1.14
     творо
    -1.07
     quando
    -1.00
    nehå
    -0.97
    prio
    -0.97
     enjo
    -0.96
     जब
    -0.96
    镊
    -0.96
     ejecut
    -0.95
    carnated
    -0.94
    POSITIVE LOGITS
     evaluation
    1.69
     test
    1.63
     criteria
    1.48
     testing
    1.42
     assessment
    1.38
     tests
    1.34
     evaluating
    1.32
     evaluations
    1.31
     decision
    1.31
     screening
    1.26
    Activations Density 0.048%

    No Known Activations